Pancreatic cancer is the fourth most-common cause of

Special Section—HSCP 2014 Spring Symposium, Part II Difficult Diagnostic Problems in Pancreatobiliary Neoplasia Jacob R. Bledsoe, MD; Shweta A. Shina...
Author: Daniel Matthews
1 downloads 0 Views 3MB Size
Special Section—HSCP 2014 Spring Symposium, Part II

Difficult Diagnostic Problems in Pancreatobiliary Neoplasia Jacob R. Bledsoe, MD; Shweta A. Shinagare, MD; Vikram Deshpande, MD

 Context.—Many common diagnostic dilemmas are encountered in pancreatobiliary pathology, frequently resulting in uncertainty on behalf of the pathologist and referral for a second opinion. Objectives.—To review 4 common diagnostic dilemmas encountered in the practice of pancreatobiliary pathology: (1) pancreatic ductal adenocarcinoma versus chronic pancreatitis; (2) pancreatic ductal carcinoma versus adenocarcinomas arising in the ampulla and intrapancreatic common bile duct; (3) the distinction of uncommon intraductal neoplasms—intraductal oncocytic papillary neoplasm, intraductal tubulopapillary neoplasm, and intraductal acinar cell carcinoma; and (4) intrahepatic cholangiocarcinoma versus metastatic carcinoma.

Data Sources.—A review of pertinent literature, along with the authors’ personal experience, based on institutional and consultation materials. Conclusions.—Important diagnostic features for a few challenging problems in pancreatobiliary pathology are reviewed. Careful study of the microscopic features along with awareness of differential diagnoses and diagnostic pitfalls generally allows distinction of these entities. We also highlight established and novel ancillary studies that help to arrive at an accurate diagnosis. (Arch Pathol Lab Med. 2015;139:848–857; doi: 10.5858/ arpa.2014-0205-RA)

P

ancreatic cancer is the fourth most-common cause of cancer-related death in the United States, with an estimated 46 420 new cases and 39 590 deaths in 2014.1 Cancers of the extrahepatic biliary tract and gallbladder are relatively less common, with an estimated 10 650 new cases in 2014.1 Pancreatobiliary carcinomas have a notoriously poor prognosis with an overall 5-year survival rate of less than 5% for pancreatic ductal adenocarcinoma (PDAC),2–4 and few of these tumors are resectable at presentation because of the concomitant presence of metastasis or involvement of regional vital structures, such as the superior mesenteric or portal vasculature. In those pancreatic tumors that are resectable, the clinical and radiologic presentation demonstrates a high degree of overlap with a number of other neoplastic and pseudoneoplastic processes that can confound the clinical diagnosis. Furthermore, the histologic differentiation of pancreatic neoplasms from each other, from extrapancreatic tumors involving the pancreas, and even from benign processes, can be a challenge for the pathologist, especially on small biopsies or frozen sections. Nevertheless, careful correlation of clinical presentation,

gross pathology, and histopathologic findings allows distinction in most cases. Arguably, the most-common diagnostic problem encountered in the pancreas by the practicing pathologist is distinguishing PDAC from chronic pancreatitis and from extrapancreatic carcinomas involving the pancreas, such as distal common bile duct or ampullary carcinomas. A lesscommon diagnostic dilemma in the pancreas, but one that is potentially underrecognized, is the differentiation among nonclassic intraductal pancreatic neoplasms, specifically intraductal tubulopapillary neoplasms (ITPNs), intraductal oncocytic papillary neoplasms (IOPNs), and acinar cell carcinoma with intraductal growth. Outside the pancreas, difficult diagnostic predicaments are often encountered with intrahepatic biliary proliferations, such as the distinction of intrahepatic cholangiocarcinoma from metastases to the liver. This review presents a brief discussion of these common diagnostic dilemmas with an overview of each entity and tools that may be useful to avoid some of the diagnostic pitfalls encountered in the course of routine surgical pathology sign out.

Accepted for publication July 10, 2014. From the Department of Pathology, Massachusetts General Hospital, Boston (Drs Bledsoe and Deshpande); and the Department of Pathology and Laboratory Medicine, Tufts Medical Center, Boston (Dr Shinagare). Dr Deshpande is a coprimary investigator in a sponsored research agreement between Affymetrix, Inc, and Massachusetts General Hospital. The other authors have no relevant financial interest in the products or companies described in this article. Reprints: Vikram Deshpande, MD, Department of Pathology, Massachusetts General Hospital, Warren Bldg 2–256, 55 Fruit St, Boston, MA 02114 (e-mail: [email protected]).

PANCREATIC DUCTAL ADENOCARCINOMA VERSUS CHRONIC PANCREATITIS Nonmalignant pathologies, such as chronic pancreatitis, including special variants thereof (eg, autoimmune pancreatitis and groove pancreatitis), may show overlapping features with pancreatic carcinoma both clinically and radiologically and, occasionally, may be indistinguishable from PDAC.5,6 Because only a small proportion of radiologically detected masses in the pancreas are benign, preoperative pathologic sampling is occasionally omitted

848 Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al

Table 1. Helpful Diagnostic Features of Chronic Pancreatitis and Pancreatic Ductal Adenocarcinoma

Feature Desmoplastic stroma Irregular ducts with incomplete lumina Nuclear variation of .4:1 within a gland Cytoplasm Lobular architecture maintained Nonlobular distribution of glands Association of ducts with arteries Perineural invasion

Chronic Pancreatitis

Pancreatic Ductal Adenocarcinoma

No No

Yes Yes

No

Yes

Dense eosinophilic Yes

Pale No

No

Yes

No

Yes

No

Yes

and surgery performed without a tissue diagnosis. Therefore, pathologists may encounter pancreatic resection specimens that fail to harbor malignancy, much to the surprise of the surgeon. The histologic overlap between PDAC and chronic pancreatitis often further complicates the issue. By gross examination, ductal adenocarcinomas are typically poorly circumscribed, solid, firm masses that are difficult to distinguish from surrounding fibrotic tissue. Although usually located in the head of the pancreas, approximately 15% arise from the body or tail, and a similar proportion diffusely involve the pancreas.7 The cut surface is typically white-tan and fibrotic appearing and, occasionally, contains areas of necrosis. Rarely, the tumors undergo cystic degeneration or are associated with cystic dilation of the ducts and may, therefore, mimic primary cystic neoplasms of the pancreas. The tumors often involve the distal common bile duct and main pancreatic duct, classically resulting in abrupt stenosis with upstream duct dilation. Similarly, involvement of the ampulla or wall of the duodenum occurs not infrequently and may suggest an extrapancreatic carcinoma invading the pancreas (discussed below). A thoughtful dissection of the specimen will generally locate the epicenter of the tumor in the head of the pancreas, a finding that argues for a primary pancreatic carcinoma. In contrast, chronic pancreatitis is grossly characterized by uneven and poorly circumscribed fibrosis of the parenchyma, which usually involves the pancreas more diffusely than it does in PDAC and is often associated with calculi in the ducts and pseudocyst formations. Other forms of pancreatitis, such as autoimmune pancreatitis and paraduodenal pancreatitis, may be associated with a discrete mass. Stenosis of the common bile duct and pancreatic duct may occur in chronic pancreatitis but is typically tapering rather than abrupt.8 Useful histologic features for establishing the malignant nature of pancreatic lesions are well documented (Table 1)9 and include poorly circumscribed, irregular, or angulated malignant glands associated with abundant desmoplastic stroma and ducts outside of their normal lobular distribution (Figure 1, A and B). The presence of ducts at an aberrant location constitutes an important diagnostic parameter— normally only a few medium-to-large caliber ducts should be found in the interlobular septa. Thus, the presence of ducts adjacent to an artery (unlike in the liver where ducts Arch Pathol Lab Med—Vol 139, July 2015

accompany arteries) or at a perineural location is highly suggestive of malignancy (Figure 1, C and D).9,10 Other helpful parameters for a diagnosis of malignancy include incomplete duct lumina, the presence of intraluminal mucin or necrotic glandular debris, and single-cell infiltration (Figure 1, E).9,11 The architectural features are often emphasized at the expense of cytologic parameters, primarily because, in many cases, the cells have a rather banal appearance. Cytologically, the 2 most-reliable nuclear features in PDAC are markedly irregular nuclear outlines and anisonucleosis (.4:1 variation in nuclear size within a single gland).11 Biomarkers of Pancreatic Ductal Adenocarcinoma There are a host of markers that may be of assistance in distinguishing benign from malignant pancreatic glands. Some of these, such as mucins, including Muc4 and Muc16, mesothelin, and clusterin-b, are either not widely available or have not been rigorously validated.12,13 A number of markers, for example mucin-1 (MUC1) and mesothelin, among many others, may be useful for prognostication in PDAC, but their utility for diagnosis is not well studied.14,15 Arguably, the most robust immunohistochemical marker of PDAC is SMAD4. The protein is lost in approximately onehalf of all PDACs,16 whereas nuclear reactivity is preserved in reactive and inflammatory diseases of the pancreas, such as chronic pancreatitis. Total loss of protein expression in PDAC results from biallelic mutations or mutation in one copy of the gene followed by either deletion or epigenetic silencing of the second copy. There is a strong correlation between SMAD4 mutations and loss of immunoreactivity, and thus, immunohistochemistry serves as a robust surrogate marker for the mutation (Figure 1, F). When evaluating this stain, adjacent benign tissue, such as stroma and inflammatory cells, should be regarded as an internal control. Interestingly, the SMAD4 wild-type tumor has been shown to have a better prognosis when compared to tumors with SMAD4 mutations.17 Finally, a number of genetic mutations are also observed in pancreatic carcinomas, which distinguish them from benign processes (Table 2) but not necessarily from noninvasive pancreatic neoplasia, such as pancreatic intraepithelial neoplasia.18 The most frequent among those genetic mutations are KRAS, TP53, SMAD4 and CDKN2A/p16.19,20 Virtually all pancreatic carcinomas will have one or more of these mutations. Chronic Pancreatitis In benign conditions such as chronic pancreatitis, other than an occasional large duct, few glands should be seen in the interlobular septae (Figure 2, A). Glands located within lobules, regardless of the degree of atypia, would suggest a ‘‘benign’’ interpretation. It should be remembered that a high percentage of pancreata, particularly those in elderly individuals, can show intrapancreatic fat, and thus, the presence of atypical ducts within fat is not diagnostic for carcinoma and does not constitute extrapancreatic invasion—a detail that is relevant in the staging of pancreatic carcinoma, specifically the distinction of pT2 from pT3 tumors. In chronic pancreatitis, reactive nuclear atypia is seen, but variation in nuclear size should be minimal. Benign and reactive ducts typically show dense eosinophilic cytoplasm (Figure 2, B). In contrast, many adenocarcinomas, especially well-differentiated carcinomas, show abundant, pale, apical cytoplasm (Figure 1, E), which gives the cells a low nuclear to cytoplasmic ratio, a Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al 849

Figure 1. Pancreatic ductal adenocarcinoma. A, Atypical angulated ducts with abundant surrounding desmoplastic stroma. B, Haphazardly organized glands outside of the normal ductal distribution, within fat and fibrous tissue adjacent to a large muscular vessel. C, The presence of isolated glands adjacent to a medium-sized artery on frozen section is highly suggestive of an invasive carcinoma. D, Perineural invasion is a useful diagnostic feature of pancreatic ductal adenocarcinoma, particularly in cases with deceptively benign-appearing malignant glands. E, Conventional features of pancreatic ductal carcinoma include angulated glands with incomplete lumina, pale cytoplasm, intraluminal necrotic debris, and surrounding desmoplastic stroma. F, Loss of SMAD4/DPC4 expression in atypical glands is the most useful biomarker for pancreatic ductal adenocarcinoma. Note that the islets provide a robust positive internal control (hematoxylin-eosin, original magnifications 3100 [A and B], 3200 [C], 3400 [C inset, D, and E]; SMAD4, original magnification 3400 [F]).

850 Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al

Table 2.

Gene

Genetic Markers that Distinguish Benign and Malignant Pancreatic Lesions

Role in Tumorigenesis

Pancreatic Cancers Showing Alteration, %

Mechanisms of Activation/ Inactivation

KRAS

Oncogene

TP53

Tumor suppressor gene

50

Mutation in one allele and loss of other allele

SMAD4

Tumor suppressor gene

50

Homozygous deletion or mutation in one allele and loss of other allele

CDKN2A/p16

Tumor suppressor gene

95

Homozygous deletion or mutation in one allele and loss of other allele or promoter methylation

.90

Activating mutation in codons 12, 13, or 61

Diagnostic Value May provide supportive evidence for pancreatic adenocarcinoma and intraductal papillary mucinous neoplasms Caveat: Virtually all pancreatic intraepithelial neoplasia lesions also harbor KRAS mutation Strong reactivity for p53 is seen in 50% of PDACs Caveat: Immunoreactivity for p53 does not always correlate with p53 mutation, ie, p53 reactivity may be seen in reactive lesions On immunohistochemistry, loss of SMAD4 is diagnostic for PDAC Caveat: Technically difficult stain, ensure positive control on tissue before reporting Not helpful diagnostically

Abbreviation: PDAC, pancreatic ductal adenocarcinoma.

variants of chronic pancreatitis, the strong clinical suspicion for malignancy may bias the pathologist into suspecting a well-differentiated adenocarcinoma.

common finding in PDAC and one that can create a deceptively bland appearance. Islets can be especially problematic mimics of carcinoma in chronic pancreatitis. Typically, islets are relatively unaffected by the inflammatory process and appear as small, benign-appearing nests of cells embedded in fibrotic stroma. However, in pancreatitis islets may become hyperplastic or diffuse and can be closely associated with nerves and muscular vessels, mimicking endocrine neoplasia or poorly differentiated carcinoma, particularly on frozen section. The distinction of autoimmune and paraduodenal pancreatitis from PDAC may be especially difficult clinically and radiologically but is usually straight forward histologically. Details of these entities have been previously described in depth.21,22 The presence of storiform-type fibrosis and obliterative phlebitis would indicate type 1 autoimmune pancreatitis, whereas a periductal infiltrate accompanied by neutrophilic abscesses would support type 2 autoimmune pancreatitis. The key ancillary test for the diagnosis of type 1 autoimmune pancreatitis is an immunohistochemical assay for immunoglobulin G4 (IgG4). Notably, approximately 10% of pancreatic adenocarcinomas are associated with increased IgG4þ plasma cells,23 so correlation with histomorphology is required. In paraduodenal, or groove, pancreatitis, the mass lesion is located in the groove between the duodenum, the common bile duct, and the pancreatic head, adjacent to the minor papilla. Fibrosis involves both the duodenal wall and the adjacent pancreas.22 Embedded within the lesion are cystic spaces lined by granulation tissue, prompting the designation cystic dystrophy of heterotopic pancreas, or paraduodenal wall cyst,24 and exuberant Brunner gland hyperplasia is invariably seen.25 Although this form of pancreatitis may mimic pancreatic carcinoma clinically, the histologic features are generally unequivocally benign and the few ‘‘atypical’’ pancreatic ducts embedded within the lesion do not raise a concern for malignancy. However, similar to other mass-forming

Grossly, identification of the epicenter of the tumor in an extrapancreatic location such as the ampulla strongly suggests a nonpancreatic malignancy (Figure 3, A). Most pathologists refer to tumors arising from the ampullary papilla, the distal-most common bile duct and the pancreatic duct, and the common channel as ampullary carcinomas. Identification of the pattern of ampullary involvement is often possible and prognostically relevant categorization of ampullary carcinomas has recently been established.26 These 4 categories of ampullary carcinoma

Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al 851

PANCREATIC DUCTAL ADENOCARCINOMA VERSUS DISTAL BILE DUCT AND AMPULLARY CARCINOMA The ampullary region is anatomically and histologically complex because it serves as the confluence of the common bile duct, pancreatic duct, duodenum, and pancreas. Therefore, adenocarcinomas arising from these extrapancreatic structures and invading the pancreas may be difficult to distinguish from PDAC, although the latter is encountered considerably more often. Although the distinction is not easily made on a biopsy sample, the distinction is obligatory on a Whipple resection because each site has a specific TNM staging system, and the site of origin may affect the choice of adjuvant therapy and entry into clinical trials. An accurate diagnosis is contingent upon identification of the primary site of tumor involvement, which is accomplished through a proper gross dissection in which sections are taken that demonstrate the relationship between tumor, ampulla, periampullary duodenum, distal common bile duct, and pancreas. We find that is best accomplished through probing and opening anteriorly along both the common bile duct and pancreatic duct. Ampullary Carcinoma Versus Pancreatic Ductal Adenocarcinoma

Figure 2. Chronic pancreatitis. A, Although atypical glands can be seen in chronic pancreatitis, they are limited to a normal lobular distribution and do not extend into interlobular fibrous septae. B, Highpower view of the A inset. In comparison to pancreatic ductal adenocarcinoma, reactive ducts show complete lumina, more uniformity in nuclear size, open chromatin, and relatively dense eosinophilic cytoplasm (hematoxylin-eosin, original magnifications 3100 [A] and 3400 [B]).

include (1) intra-ampullary papillary-tubular neoplasm with invasion, in which a predominant preinvasive component is present in the common channel of the ampulla; (2) ampullary-ductal type carcinoma, which is usually a pancreatobiliary type and shows invasion associated with fibrous thickening of the ampullary duct; (3) periampullaryduodenal carcinoma, which shows prominent exophytic growth on the duodenal surface of the papilla with involvement of the ampullary orifice but a relatively small intra-ampullary component; and (4) ampullary-not otherwise specified (papilla of Vater), in which the carcinoma is localized at the tip of the ampullary papilla, often with ulceration.26 Among these groups, the ampullary-ductal type was associated with the poorest overall survival, but all subtypes had a better prognosis than PDAC, therefore, necessitating accurate diagnosis.26 Besides prognostication, use of this classification scheme may decrease confusion about the spectrum of tumors that should be designated ampullary carcinoma. 852 Arch Pathol Lab Med—Vol 139, July 2015

Similar to nonampullary duodenal carcinomas, ampullary adenocarcinomas most commonly have an intestinal-type morphology characterized by atypical, pseudostratified, columnar epithelium with elongated nuclei forming glands reminiscent of colorectal carcinoma (Figure 3, B). In contrast, pancreatobiliary-type tumors form small, infiltrative glands with 1 to 2 cuboidal cell layers (Figure 3, C). Ampullary carcinoma with a pancreatobiliary or combined intestinal/biliary phenotype can be troublesome to differentiate from PDAC because of morphologic overlap. Histologically, the most helpful clue pointing to an ampullary origin is the presence of a preinvasive component within the most distal portion of the bile duct or ampulla; such lesions frequently have a tubulopapillary appearance.27 Intestinal-type ampullary and nonampullary duodenal adenocarcinomas usually express keratin 20, CDX2, and MUC2 but are negative for MUC1, MUC5AC, and keratin 7 in most cases.28,29 Pancreatobiliary-type carcinomas of the ampulla usually have the opposite staining pattern, which is shared with PDAC.29 Among these markers, CDX2, MUC1, and MUC2 are particularly reliable, and the keratin stains are relatively unreliable.30 However, these markers may be less effective in distinguishing among phenotypes in ampullary neoplasia than in pancreatic intraductal papillary mucinous neoplasms (IPMNs) given the mix of cell types normally found at the ampulla.27 Notably, cancerization of the ampullary, bile ductular, and duodenal epithelium by PDAC may occur and can be difficult to distinguish from primary carcinoma in situ. The aforementioned immunostains may be of use in distinguishing such colonization from a primary dysplastic process of these sites.29,31 Like PDAC, loss of DPC4 (SMAD4) expression has been documented in ampullary carcinomas, and expression of CEA and CA 19-9 is often seen and cannot be used to differentiate these tumors.32 Given the prominent fibrotic response and poor circumscription that can be seen in tumors arising at or near the ampulla, gross and histopathologic distinction of the primary site is not always possible in large tumors. At a minimum, documentation of differentiation (intestinal, pancreatobiliary, or mixed) should be reported because studies have shown that intestinal-type ampullary carcinomas carry a better prognosis than do pancreatobiliary-type tumors.30,33–35 However, the optimal method of distinguishing the intestinal from the pancreatobiliary phenotypes remains to be determined—through morphology alone or through a combination of morphology and immunohistochemistry. A recent analysis36 used a combination of CDX2, MUC1, and morphology to separate pancreatobiliary (MUC1þ CDX2) from intestinal (MUC1 CDX2þ) tumors and showed that the phenotype, particularly when combined with lymph node status, was highly predictive of prognosis. Although pure forms of intestinal and pancreatobiliary phenotypes exist, many cases show elements of both. Finally, it may be difficult or impossible to determine the phenotype of non-tubule forming variants, such as medullary or signet ring cell carcinomas. Distal Bile Duct Carcinoma Versus Pancreatic Ductal Adenocarcinoma Distinction of PDAC from distal bile duct carcinoma involving the pancreas may be even more problematic than distinguishing ampullary tumors because of their intrapancreatic location and identical pancreatobiliary-type morphology and immunophenotype. The distinction between Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al

Figure 3. Ampullary and distal bile duct carcinomas. A, Gross photo of an ampullary carcinoma (arrow). The relative location of the tumor is apparent after opening the duodenum, common bile duct (CBD), and the pancreatic duct. B, Intestinal-type ampullary carcinoma with large glands lined by atypical, pseudostratified columnar cells reminiscent of colorectal carcinoma. C, Ampullary carcinoma with a pancreatobiliary phenotype has an identical appearance to pancreatic ductal adenocarcinoma and distal bile duct carcinoma. D, Concentric involvement of the bile duct is a useful diagnostic feature of distal bile duct carcinoma, which otherwise looks identical to pancreatic ductal carcinoma (hematoxylin-eosin, original magnifications 3400 [B and C], 340 [D], and 3200 [D inset]).

these entities may be important for stratification of patients into clinical trials or treatment groups and has prognostic significance.33 Distal bile duct carcinomas are frequently misclassified as either PDAC or ampullary carcinoma, and survival in cases reclassified from PDAC is better than it is in true PDAC.37 The distinction may also have therapeutic implications; in one study,38 treatment of patients who have bile duct carcinoma (including advanced or metastatic cholangiocarcinoma, gallbladder cancer, and ampullary cancer) with cisplatin plus gemcitabine (the latter being a key agent used to treat PDAC) was associated with a significant survival advantage as compared with use of gemcitabine alone, without the addition of substantial toxicity. Helpful features that point to a distal bile duct origin include the finding of an in situ component, such as prominent biliary intraepithelial neoplasia or a biliary intraductal papillary neoplasm, as well as circumferential involvement of the bile duct by invasive carcinoma (Figure 3, D). In tumors situated near the bile duct, circumferential and symmetric involvement was found to be associated with

UNCOMMON PANCREATIC INTRADUCTAL PAPILLARY LESIONS Intraductal papillary mucinous neoplasms are often seen in clinical practice, and although they may cause some diagnostic dilemmas, such as the differential with mucinous cystic neoplasms, the characteristic features of classic gastric-type, intestinal-type, and pancreatobiliary-type IPMNs are generally widely known and have been reviewed previously.40,41 In contrast, the diagnosis of oncocytic-type

Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al 853

bile duct dysplasia, whereas tumors with their epicenter away from the common bile duct were more likely associated with pancreatic intraepithelial neoplasia, suggesting a pancreatic origin.39 Therefore, when a tumor is centered in the bile duct, we are inclined to favor the diagnosis of bile duct carcinoma. Otherwise, we are hesitant to make a diagnosis of distal bile duct carcinoma, regardless of the extent of bile duct involvement. Currently, other than careful gross and histologic examination, reliable markers that can distinguish distal bile duct carcinoma from PDAC are lacking.

Table 3. Comparative Immunohistochemistry of Intraductal Oncocytic Papillary Neoplasms, Intraductal Tubulopapillary Neoplasms, and Intraductal Acinar Cell Carcinoma Immunohistochemistry Immunohistochemistrya Tumor

Acinar Growth CK7 CK19 MUC1 MUC2 MUC5AC MUC6 CDX2 Trypsin Chymotrypsin

DPC4/ SMAD4

Other b

Mitochondrial stains: phosphotungstic acid hematoxylin, Novelli stain, 111.3 antibody42

IOPN



þ

þ

þ

focal/

focal

þ







Retained

ITPN IACC

þ

þ 

þ 

6 

 

 

6 

 

 þ

 þ

Retainedb Retainedb Other exocrine markers

þ

Abbreviations: IACC, intraductal acinar cell carcinoma; IOPN, intraductal oncocytic papillary neoplasm; ITPN, intraductal tubulopapillary neoplasm. a Typical staining patterns based on the authors’ experience. Staining defined as typically positive (þ), typically negative (), variable (6), typically focally positive (focal), or typically negative or only focally positive (focal/). b Loss of DPC/SMAD4 expression can be seen in a corresponding invasive component.

IPMNs, such as IOPNs and other infrequently encountered intraductal papillary lesions, such as intraductal tubulopapillary neoplasms and acinar cell carcinoma with intraductal growth, may cause more of a diagnostic dilemma given their degree of morphologic overlap and most pathologists’ lack of familiarity with these entities. Knowledge of the typical histology and characteristic immunohistochemical profiles of these lesions (Table 3) is helpful. Intraductal Oncocytic Papillary Neoplasm Intraductal oncocytic papillary neoplasms, like the other types of IPMNs, are often grossly cystic with mucin production and duct dilation. Histologically, they are characterized by a multilayered proliferation of atypical oncocytic cells with large nuclei, prominent nucleoli, and distinctive eosinophilic cytoplasm (Figure 4, A and B). Intraductal oncocytic papillary neoplasms show complex arborizing papillae, cribriform formations, and solid nests within the lumen of a dilated duct. Mucin-containing intraepithelial and intracellular lumina, as well as scattered goblet cells, may be seen (Figure 4, B).42,43 These neoplasms show diffuse high-grade dysplasia and may evolve into invasive carcinoma with similar oncocytic features. Intraductal oncocytic papillary neoplasms are typically positive for MUC6 (Figure 4, C) and MUC1, may be focally positive for MUC5AC, and are usually negative for MUC2 and CDX2, except in interspersed goblet cells.43,44 Intraductal oncocytic papillary neoplasms characteristically lack the KRAS mutations that are found in PDAC and many nononcocytic IPMNs.45,46 Intraductal Tubulopapillary Neoplasm Intraductal tubulopapillary neoplasms are relatively rare, making up approximately 3% of the intraductal neoplasms of the pancreas, and they are clinically and histopathologically distinct from IPMNs.27 Radiologically and grossly, they form solid masses with no mucin production. Histologically, the intraductal nodules are predominantly made up of closely apposed, tubule-forming glands and cribriform structures with uniform high-grade dysplasia (Figure 4, D and E).47 There is morphologic overlap with the pancreatobiliary type of IPMNs, but the latter do not show complex tubulopapillary proliferations expanding and clogging the pancreatic duct, which are characteristic of ITPNs. The cuboidal cells show a modest amount of eosinophilic to 854 Arch Pathol Lab Med—Vol 139, July 2015

amphophilic cytoplasm (Figure 4, E). Mitoses are frequently seen, and necrosis is more commonly seen than is seen in IPMNs.48 In contrast to IPMNs, mucin is absent, and the broad range of atypia is typically not seen. An associated invasive tubular carcinoma is often present.48 Stains for mucins, such as MUC1 and MUC6, can be positive but are more often negative than they are in IPMNs, including IOPN, and trypsin and chymotrypsin staining should be absent (Figure 4, F). Unlike IPMNs, these tumors are negative for KRAS mutations and, instead, show mutations in the PI3K pathway.48,49 Intraductal Acinar Cell Carcinoma Acinar cell carcinomas occasionally have a component of intraductal polypoid growth,50 and when predominantly intraductal, they can mimic other intraductal neoplasms, such as ITPNs and IOPNs.51,52 Grossly, such tumors show a nodular or polypoid intraductal growth pattern accompanied by duct dilation and surrounding parenchymal sclerosis. Like ITPNs, acinar cell carcinomas are typically composed of sheets of back-to-back acinar structures (Figure 4, G and H). True tumor-lining papillary structures may be seen.53 The nuclei are round, relatively uniform, and typically contain a single prominent central nucleolus (Figure 4, H). The presence of PASþ, diastase-resistant, apical, eosinophilic zymogen granules and intraluminal concretions are helpful features in distinguishing acinar cell carcinomas from other intraductal lesions but are not always present.53 Immunohistochemistry for trypsin (Figure 4, I) and chymotrypsin, both of which are positive in acinar cell carcinoma, may be invaluable for distinguishing intraductal acinar cell carcinoma from other intraductal lesions, such as ITPNs, which can appear quite similar on hematoxylin-eosin. The absence of keratin 19 (positive in virtually all ductal neoplasms) may be the first clue to an acinar cell proliferation. Other neoplasms that may show a prominent intraductal pattern of growth include pancreatic endocrine neoplasms54 and invasive pancreatic adenocarcinomas, which may show a cystic papillary pattern of growth, mimicking an IPMN.55,56 INTRAHEPATIC CHOLANGIOCARCINOMA VERSUS CARCINOMA METASTATIC TO THE LIVER The differentiation of an intrahepatic cholangiocarcinoma (IHCC) from a metastatic tumor of unknown primary constitutes a major unsolved problem in diagnostic surgical Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al

Figure 4. Uncommon intraductal neoplasms. A through C, Intraductal oncocytic papillary neoplasm consisting of papillae lined by oncocytic cells with prominent nucleoli projecting into the lumen of the duct. Characteristic intraepithelial and intracellular lumen formation is sometimes seen (B). Immunostains for MUC6 (C) and MUC1 are typically positive. D through F, Intraductal tubulopapillary neoplasm. Complex tubular and cribriform growth filling the duct lumen. Distinction from intraductal oncocytic papillary neoplasm and acinar cell carcinoma can be made based on hematoxylin-eosin morphology and immunohistochemistry. Intraductal tubulopapillary neoplasms should be negative for trypsin (F) and chymotrypsin. G through I, Acinar cell carcinoma with prominent intraductal growth. Acinar morphology is almost always apparent, but a papillary growth pattern may be seen (G). Apical eosinophilic granularity may be appreciable on high power. Immunostains for exocrine proteases trypsin (I) and chymotrypsin are positive (hematoxylin-eosin, original magnifications 3200 [A, D, and G] and 3400 [B, E, and H]; original magnification 3400 [C, F, and I]).

pathology. The morphologic appearance is of some value in this distinction. Most IHCCs are composed of well-formed, round to oval glands lined by low cuboidal epithelium that is devoid of mucin. The tumor frequently displays a ‘‘neverending’’ glandular pattern in which the glands appear to merge into each other without intervening stroma (Figure 5, A). A substantial proportion of IHCC cases have a bland morphology and thus often bear a superficial resemblance to bile duct adenomas. Other common patterns of growth include solid nests and glands with cribriform structures. Focal areas of clear cell change are commonly observed (Figure 5, B). The glands are present in the abundant stroma, which is often hyalinized. The IHCC cases tend to differ morphologically from perihilar and distal bile duct carcinomas, which instead resemble PDACs. Despite these characteristics, the poorly differentiated variant of IHCC cannot be distinguished with confidence from a metastatic carcinoma by histology or immunohistochemistry. That

conundrum can result in an extensive clinical workup to exclude a potential tumor of nonhepatic origin. Although immunohistochemical confirmation is required for a definitive diagnosis of IHCC, there are no markers unique to cholangiocarcinoma, and the foremost role of an immunohistochemical panel is to exclude a metastatic adenocarcinoma. However, the keratin profile is identical to a host of other tumors: keratin 7þ and 19þ, and keratin 20. Access to a definitive marker for IHCC would allow the oncologist to forego the obligatory, and often extensive, immunohistochemical workup. The introduction of 2 markers may change the current diagnostic algorithm in IHCC: albumin and IDH1/2. Recent data from our group suggest that the presence of albumin in IHCC, detected using branch chain chromogenic in situ hybridization technology (ViewRNA, Affymetrix, Santa Clara, California), can distinguish this neoplasm from metastatic adenocarcinoma (Figure 5, C and D).57 Thus, detecting albumin in a gland-forming tumor that lacks histologic evidence of hepatocytic differentiation is

Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al 855

Figure 5. Intrahepatic cholangiocarcinoma. A, Intrahepatic cholangiocarcinoma is typically composed of infiltrating tubular glands with a ‘‘neverending’’ pattern, in which one gland merges imperceptibly with another in the setting of background fibrosis. B, Solid nests with cytoplasmic clearing are frequently seen in intrahepatic cholangiocarcinoma. C and D, Chromogenic in situ hybridization for albumin is diffusely positive in the tumors depicted in A and B (hematoxylin-eosin, original magnification 3200 [A and B]; original magnification 3200 [C and D]; original magnification 3400 [D inset]).

evidence of IHCC.57 Similarly the presence of IDH1/2 mutations in a gland-forming tumor in the liver is highly suggestive of IHCC, although such mutations are present in only one-third of these neoplasms.58 IDH1/2 mutations are identified in a variety of other tumor types, such as glioma, chondrosarcoma, and acute myeloid leukemia; however, none of those tumors are likely to be confused with IHCC. References 1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin. 2014;64(1):9–29. 2. Hidalgo M. Pancreatic cancer. N Engl J Med. 2010;362(17):1605–1617. 3. Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet. 2004; 363(9414):1049–1057. 4. Maitra A, Hruban RH. Pancreatic Cancer. Annu Rev Pathol. 2008;3(1):157– 188. 5. Tamm EP, Bhosale PR, Vikram R, de Almeida Marcal LP, Balachandran A. Imaging of pancreatic ductal adenocarcinoma: state of the art. World J Radiol. 2013;5(3):98–105. 6. Tempero MA, Arnoletti JP, Behrman S, et al; NCCN Pancreatic Adenocarcinoma. Pancreatic adenocarcinoma. J Natl Compr Canc Netw. 2010;8(9):972– 1017. 7. Hruban RH, Pitman MB, Klimstra DS. Tumors of the Pancreas. Washington, DC: American Registry of Pathology and Armed Forces Institute of Pathology; 2007. AFIP Atlas of Tumor Pathology; 4th series, fascicle 6. 8. Kloppel G, Adsay NV. Chronic pancreatitis and the differential diagnosis ¨ versus pancreatic cancer. Arch Pathol Lab Med. 2009;133(3):382–387.

856 Arch Pathol Lab Med—Vol 139, July 2015

9. Hyland C, Kheir SM, Kashlan MB. Frozen section diagnosis of pancreatic carcinoma: a prospective study of 64 biopsies. Am J Surg Pathol. 1981;5(2):179– 191. 10. Sharma S, Green KB. The pancreatic duct and its arteriovenous relationship: an underutilized aid in the diagnosis and distinction of pancreatic adenocarcinoma from pancreatic intraepithelial neoplasia: a study of 126 pancreatectomy specimens. Am J Surg Pathol. 2004;28(5):613–620. 11. Cioc AM, Ellison EC, Proca DM, Lucas JG, Frankel WL. Frozen section diagnosis of pancreatic lesions. Arch Pathol Lab Med. 2002;126(10):1169–1173. 12. Jhala N, Jhala D, Vickers SM, et al. Biomarkers in diagnosis of pancreatic carcinoma in fine-needle aspirates. Am J Clin Pathol. 2006;126(4):572–579. 13. Shimizu A, Hirono S, Tani M, et al. Coexpression of MUC16 and mesothelin is related to the invasion process in pancreatic ductal adenocarcinoma. Cancer Sci. 2012;103(4):739–746. 14. Winter JM, Tang LH, Klimstra DS, et al. A novel survival-based tissue microarray of pancreatic cancer validates MUC1 and mesothelin as biomarkers. PLoS One. 2012;7(7):e40157. doi:10.1371/journal.pone.0040157. 15. Smith RA, Tang J, Tudur-Smith C, Neoptolemos JP, Ghaneh P. Meta-analysis of immunohistochemical prognostic markers in resected pancreatic cancer. Br J Cancer. 2011;104(9):1440–1451. 16. Tascilar M, Skinner HG, Rosty C, et al. The SMAD4 protein and prognosis of pancreatic ductal adenocarcinoma. Clin Cancer Res. 2001;7(12):4115–4121. 17. Blackford A, Serrano OK, Wolfgang CL, et al. SMAD4 gene mutations are associated with poor prognosis in pancreatic cancer. Clin Cancer Res. 2009; 15(14):4674–4679. 18. Feldmann G, Beaty R, Hruban RH, Maitra A. Molecular genetics of pancreatic intraepithelial neoplasia. J Hepatobiliary Pancreat Surg. 2007;14(3): 224–232.

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al

19. Vincent A, Herman J, Schulick R, Hruban RH, Goggins M. Pancreatic cancer. Lancet. 2011;378(9791):607–620. 20. Goggins M, Schutte M, Lu J, et al. Germline BRCA2 gene mutations in patients with apparently sporadic pancreatic carcinomas. Cancer Res. 1996; 56(23):5360–5364. 21. Kloppel G. Chronic pancreatitis, pseudotumors and other tumor-like ¨ lesions. Mod Pathol. 2007;20(1)(suppl):S113–S131. doi:10.1038/modpathol. 3800690. 22. Zamboni G, Capelli P, Scarpa A, et al. Nonneoplastic mimickers of pancreatic neoplasms. Arch Pathol Lab Med. 2009;133(3):439–453. 23. Hirano K, Komatsu Y, Yamamoto N, et al. Pancreatic mass lesions associated with raised concentration of IgG4. Am J Gastroenterol. 2004;99(10): 2038–2040. 24. Adsay NV, Zamboni G. Paraduodenal pancreatitis: a clinico-pathologically distinct entity unifying ‘‘cystic dystrophy of heterotopic pancreas,’’ ‘‘paraduodenal wall cyst,’’ and ‘‘groove pancreatitis.’’ Semin Diagn Pathol. 2004; 21(4):247–254. 25. Tezuka K, Makino T, Hirai I, Kimura W. Groove pancreatitis. Dig Surg. 2010;27(2):149–152. 26. Adsay V, Ohike N, Tajiri T, et al. Ampullary region carcinomas: definition and site specific classification with delineation of four clinicopathologically and prognostically distinct subsets in an analysis of 249 cases. Am J Surg Pathol. 2012;36(11):1592–1608. 27. Ohike N, Kim GE, Tajiri T, et al. Intra-ampullary papillary-tubular neoplasm (IAPN): characterization of tumoral intraepithelial neoplasia occurring within the ampulla: a clinicopathologic analysis of 82 cases. Am J Surg Pathol. 2010;34(12):1731–1748. 28. Bosman FT, Carneiro F, Hruban RH, Theise ND, eds. WHO Classification of Tumours of the Digestive System. 4th ed. Lyon, France: IARC Press; 2010. World Health Organization Classification of Tumours; vol 3. 29. Chu PG, Schwarz RE, Lau SK, Yen Y, Weiss LM. Immunohistochemical staining in the diagnosis of pancreatobiliary and ampulla of Vater adenocarcinoma: application of CDX2, CK17, MUC1, and MUC2. Am J Surg Pathol. 2005; 29(3):359–367. 30. Kumari N, Prabha K, Singh RK, Baitha DK, Krishnani N. Intestinal and pancreatobiliary differentiation in periampullary carcinoma: the role of immunohistochemistry. Hum Pathol. 2013;44(10):2213–2219. 31. Goldstein NS, Bassi D. Cytokeratins 7, 17, and 20 reactivity in pancreatic and ampulla of Vater adenocarcinomas: percentage of positivity and distribution is affected by the cut-point threshold. Am J Clin Pathol. 2001;115(5):695–702. 32. McCarthy DM, Hruban RH, Argani P, et al. Role of the DPC4 tumor suppressor gene in adenocarcinoma of the ampulla of Vater: analysis of 140 cases. Mod Pathol. 2003;16(3):272–278. 33. Howe JR, Klimstra DS, Moccia RD, Conlon KC, Brennan MF. Factors predictive of survival in ampullary carcinoma. Ann Surg. 1998;228(1):87–94. 34. Roh YH, Kim YH, Li HW, et al. The clinicopathologic and immunohistochemical characteristics of ampulla of Vater carcinoma: the intestinal type is associated with a better prognosis. Hepatogastroenterology. 2007;54(78):1641– 1644. 35. Hansel DE, Maitra A, Lin JW, et al. Expression of the caudal-type homeodomain transcription factors CDX 1/2 and outcome in carcinomas of the ampulla of Vater. J Clin Oncol. 2005;23(9):1811–1818. 36. Chang DK, Jamieson NB, Johns AL, et al. Histomolecular phenotypes and outcome in adenocarcinoma of the ampulla of Vater. J Clin Oncol. 2013;31(10): 1348–1356. 37. Pomianowska E, Grzyb K, Westgaard A, Clausen OPF, Gladhaug IP. Reclassification of tumour origin in resected periampullary adenocarcinomas reveals underestimation of distal bile duct cancer. Eur J Surg Oncol. 2012;38(11): 1043–1050. 38. Valle J, Wasan H, Palmer DH, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;362(14):1273–1281.

39. Shen J, Konstantinidis I, Ferrone CR, Deshpande V. Intrapancreatic cholangiocarcinomas: a clinicopathological and immunohistochemical analysis [abstract 1651]. Mod Pathol. 2010;23(suppl 1):371A. doi:10.1038/modpathol. 2010.22. 40. Furukawa T, Kloppel G, Adsay NV, et al. Classification of types of ¨ intraductal papillary-mucinous neoplasm of the pancreas: a consensus study. Virchows Arch. 2005;447(5):794–799. 41. Katabi N, Klimstra DS. Intraductal papillary mucinous neoplasms of the pancreas: clinical and pathological features and diagnostic approach. J Clin Pathol. 2008;61(12):1303–1313. 42. Adsay NV, Adair CF, Heffess CS, Klimstra DS. Intraductal oncocytic papillary neoplasms of the pancreas. Am J Surg Pathol. 1996;20(8):980–994. 43. Shi C, Hruban RH. Intraductal papillary mucinous neoplasm. Hum Pathol. 2012;43(1):1–16. 44. Basturk O, Khayyata S, Klimstra DS, et al. Preferential expression of MUC6 in oncocytic and pancreatobiliary types of intraductal papillary neoplasms highlights a pyloropancreatic pathway, distinct from the intestinal pathway, in pancreatic carcinogenesis. Am J Surg Pathol. 2010;34(3):364–370. 45. Patel SA, Adams R, Goldstein M, Moskaluk CA. Genetic analysis of invasive carcinoma arising in intraductal oncocytic papillary neoplasm of the pancreas. Am J Surg Pathol. 2002;26(8):1071–1077. 46. Chung SM, Hruban RH, Iacobuzio-Donahue CA, Adsay NV, Zee SY, Klimstra DS. Analysis of molecular alterations and differentiation pathways in intraductal oncocytic papillary neoplasm of the pancreas [abstract 1283]. Mod Pathol. 2005;18(1)(suppl):277A–278A. doi:10.1038/sj.modpathol.3800918. 47. Tajiri T, Tate G, Inagaki T, et al. Intraductal tubular neoplasms of the pancreas: histogenesis and differentiation. Pancreas. 2005;30(2):115–121. 48. Yamaguchi H, Shimizu M, Ban S, et al. Intraductal tubulopapillary neoplasms of the pancreas distinct from pancreatic intraepithelial neoplasia and intraductal papillary mucinous neoplasms. Am J Surg Pathol. 2009;33(8): 1164–1172. 49. Yamaguchi H, Kuboki Y, Hatori T, et al. Somatic mutations in PIK3CA and activation of AKT in intraductal tubulopapillary neoplasms of the pancreas. Am J Surg Pathol. 2011;35(12):1812–1817. 50. Ban D, Shimada K, Sekine S, et al. Pancreatic ducts as an important route of tumor extension for acinar cell carcinoma of the pancreas. Am J Surg Pathol. 2010;34(7):1025–1036. 51. Fabre A, Sauvanet A, Flejou JF, et al. Intraductal acinar cell carcinoma of the pancreas. Virchows Arch. 2001;438(3):312–315. 52. Toll AD, Mitchell D, Yeo CJ, Hruban RH, Witkiewicz AK. Acinar cell carcinoma with prominent intraductal growth pattern: case report and review of the literature. Int J Surg Pathol. 2011;19(6):795–799. 53. Basturk O, Zamboni G, Klimstra DS, et al. Intraductal and papillary variants of acinar cell carcinomas: a new addition to the challenging differential diagnosis of intraductal neoplasms. Am J Surg Pathol. 2007;31(3):363–370. 54. Chetty R, El-Shinnawy I. Intraductal pancreatic neuroendocrine tumor. Endocr Pathol. 2009;20(4):262–266. 55. Bagci P, Andea AA, Basturk O, Jang KT, Erbarut I, Adsay V. Large duct type invasive adenocarcinoma of the pancreas with microcystic and papillary patterns: a potential microscopic mimic of non-invasive ductal neoplasia. Mod Pathol. 2012;25(3):439–448. 56. Kelly PJ, Shinagare S, Sainani N, et al. Cystic papillary pattern in pancreatic ductal adenocarcinoma: a heretofore undescribed morphologic pattern that mimics intraductal papillary mucinous carcinoma. Am J Surg Pathol. 2012;36(5): 696–701. 57. Rice-Stitt TL, Mubeen A, Shahid M, Rivera M, Deshpande V. Chromogenic in situ hybridization for albumin distinguishes intrahepatic cholangiocarcinoma from non-hepatic neoplasms [abstract 812]. Mod Pathol. 2014;27(2)(suppl): 199A. doi:10.1038/modpathol.2014.13. 58. Borger DR, Tanabe KK, Fan KC, et al. Frequent mutation of isocitrate dehydrogenase (IDH)1 and IDH2 in cholangiocarcinoma identified through broad-based tumor genotyping. Oncologist. 2012;17(1):72–79.

Arch Pathol Lab Med—Vol 139, July 2015

Diagnostic Problems in Pancreatobiliary Neoplasia—Bledsoe et al 857